j2me xmlReader的轻量级实现

作者cleverpig

版权申明:可以自由转载, 转载请保留下面的作者信息:
作者 cleverpig(http://www.matrix.org.cn/blog/cleverpig)
关键字:xml,j2me,kxml,xmlpull

众所周知，xml在j2me上的使用，kxml是首当其冲的老大。它实现了xmlpull的api，提供了很严谨的移动设备上的xml读写功能。

一、kxml的前身--xmlpull

xmlpull提供了为资源有限的环境（如j2me）和j2ee应用服务器使用的xml解析API，xml pull解析允许使用stream解析的方式。

xmlpull的特点：

*  简单的接口-解析器包含一个接口、一个异常、一个建立解析器的factory；
    * 无实现依赖-模仿JAXP的factory类和很容易的切换不同的XmlPull V1 API实现，而无需修改代码；
    * 易用性-只有唯一的next()方法用来读取下面5类事件中的一个：
          START DOCUMENT-文档起始点，这时解析器还没有准备好读取文档；
          START_TAG-解析器读取到了element声明的开始标签，就是xml文件中声明element用的"<"；
          TEXT-解析器读取到了element的内容，如element的各种属性和value；
          END_TAG-解析器读取到了element声明的结束标签，就是"/>"或者"</...>"；
          END_DOCUMENT-解析器读取到文档的结尾，解析工作完毕。

* 多功能性-由于提供了xml解析器的通用接口，所以允许多个实现和在特性和属性上的扩展
    * 良好的性能-由于提供了通用的接口，可以说也就为使用快速的xml解析器提供了条件，但是目前最快的解析器，本人还没遇到过。:-)
    * 最小化需要-设计上与j2me兼容，从而使其能工作于小型设备和建立使用微量内存的xmlPull兼容的解析器。

二、建立在良好构架上的kxml

xmlpull api提供的通用的xml内容解析功能使用方式：使用一个循环来遍历到下一个事件，并切换到一个针对这个事件的分支语句处理。但是在使用xml解析器时，一些xml element不仅包括text还有其它的element（被称为子元素或者内嵌元素）。对于这样的xml数据，解析处理被简单化到了使用xmlpull api的nextTag和nextText方法。另外，require方法也可以使用一个解析状态的参数来表示希望取得数据的event类型。以上就是kxml使用方法。

如解析下面内容的简单xml文件：

<elements>
  <text>text1</text>
  <text>text2</text>
</elements>  

解析代码片断如下：

parser.nextTag();
parser.require(XmlPullParser.START_TAG, null, "elements");

while(parser.nextTag() == XmlPullParser.START_TAG) {
  parser.require(XmlPullParser.START_TAG, null, "text");

// handle element content
   System.out.println("text content: "+ parser.nextText());

parser.require(XmlPullParser.END_TAG, null, "text");
}

parser.require(XmlPullParser.END_TAG, null, "elements");

从上面的代码看出，的确很简单明了。

三、背向kxml的原因

但是如果真正的使用kxml开发j2me程序时会发现，kxml包和其依赖的xmlpull包不仅无形中增加了程序发布时的大小，而且由于没有最好的解析器带来的负面性能损耗会随着xml文件的大小成非线性增长（还好不是指数级的）。这些成为了kxml的流行障碍。

四、轻量级的xmlReader

幸运的是kobject.org网站上提供了一个轻量级的xmlReader程序，大家可以到http://kobjects.sourceforge.net/utils/ 获取。
作为简单的xml解析器，功能上与kxml类似，但是不支持命名空间和一些传统的事件。整个jar文件大小小于5KB。

以下便是其代码：
XmlReader.java

/* Copyright (c) 2002,2003, Stefan Haustein, Oberhausen, Rhld., Germany
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to deal
 * in the Software without restriction, including without limitation the rights
 * to use, copy, modify, merge, publish, distribute, sublicense, and/or
 * sell copies of the Software, and to permit persons to whom the Software is
 * furnished to do so, subject to the following conditions:
 *
 * The  above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
 * IN THE SOFTWARE. */

package org.kobjects.xml;

import java.io.*;
import java.util.*;

/** A minimalistic XML pull parser, similar to kXML, but
    not supporting namespaces or legacy events. If you need
    support for namespaces, or access to XML comments or
    processing instructions, please use kXML(2) instead. */

public class XmlReader {

/** Return value of getType before first call to next()  */

public final static int START_DOCUMENT = 0;

/** Signal logical end of xml document */

public final static int END_DOCUMENT = 1;

/** Start tag was just read */

public final static int START_TAG = 2;

/**
     * End tag was just read
     */
    public final static int END_TAG = 3;

/** Text was just read  */
    public final static int TEXT = 4;

final static int CDSECT = 5;
    final static int ENTITY_REF = 6;

static final private String UNEXPECTED_EOF =
        "Unexpected EOF";
    static final private int LEGACY = 999;

// general

public boolean relaxed;
    private Hashtable entityMap;
    private int depth;
    private String[] elementStack = new String[4];

// source

private Reader reader;

private char[] srcBuf =
        new char[Runtime.getRuntime().freeMemory() >= 1048576
            ? 8192
            : 128];

private int srcPos;
    private int srcCount;

private boolean eof;

private int line;
    private int column;

private int peek0;
    private int peek1;

// txtbuffer

private char[] txtBuf = new char[128];
    private int txtPos;

// Event-related

private int type;
    private String text;
    private boolean isWhitespace;
    private String name;

private boolean degenerated;
    private int attributeCount;
    private String[] attributes = new String[16];

private String[] TYPES =
        {
            "Start Document",
            "End Document",
            "Start Tag",
            "End Tag",
            "Text" };

private final int read() throws IOException {

int r = peek0;
        peek0 = peek1;

if (peek0 == -1) {
            eof = true;

return r;
        }
        else if (r == '/n' || r == '/r') {
            line++;
            column = 0;
            if (r == '/r' && peek0 == '/n')
                peek0 = 0;
        }
        column++;

if (srcPos >= srcCount) {
            srcCount = reader.read(srcBuf, 0, srcBuf.length);
            if (srcCount <= 0) {
                peek1 = -1;
                return r;
            }
            srcPos = 0;
        }

peek1 = srcBuf[srcPos++];
        return r;
    }

private final void exception(String desc)
        throws IOException {
        throw new IOException(
            desc + " pos: " + getPositionDescription());
    }

private final void push(int c) {
        if (c == 0)
            return;

if (txtPos == txtBuf.length) {
            char[] bigger = new char[txtPos * 4 / 3 + 4];
            System.arraycopy(txtBuf, 0, bigger, 0, txtPos);
            txtBuf = bigger;
        }

txtBuf[txtPos++] = (char) c;
    }

private final void read(char c) throws IOException {
        if (read() != c) {
            if (relaxed) {
                if (c <= 32) {
                    skip();
                        read();
                    }
            }
            else {
                    exception("expected: '" + c + "'");
            }
        }
    }

private final void skip() throws IOException {

while (!eof && peek0 <= ' ')
            read();
    }

private final String pop(int pos) {
        String result = new String(txtBuf, pos, txtPos - pos);
        txtPos = pos;
        return result;
    }

private final String readName() throws IOException {

int pos = txtPos;
        int c = peek0;
        if ((c < 'a' || c > 'z')
            && (c < 'A' || c > 'Z')
            && c != '_'
            && c != ':'
            && !relaxed)
            exception("name expected");

do {
            push(read());
            c = peek0;
        }
        while ((c >= 'a' && c <= 'z')
            || (c >= 'A' && c <= 'Z')
            || (c >= '0' && c <= '9')
            || c == '_'
            || c == '-'
            || c == ':'
            || c == '.');

return pop(pos);
    }

private final void parseLegacy(boolean push)
        throws IOException {

String req = "";
        int term;

read(); // <
        int c = read();

if (c == '?') {
            term = '?';
        }
        else if (c == '!') {
            if (peek0 == '-') {
                req = "--";
                term = '-';
            }
            else {
                req = "DOCTYPE";

term = -1;
            }
        }
        else {
            if (c != '[')
                exception("cantreachme: " + c);
            req = "CDATA[";
            term = ']';
        }

for (int i = 0; i < req.length(); i++)
            read(req.charAt(i));

if (term == -1)
            parseDoctype();
        else {
            while (true) {
                if (eof)
                    exception(UNEXPECTED_EOF);

c = read();
                if (push)
                    push(c);

if ((term == '?' || c == term)
                    && peek0 == term
                    && peek1 == '>')
                    break;
            }
            read();
            read();

if (push && term != '?')
                pop(txtPos - 1);
        }
    }

/** precondition: &lt! consumed */

private final void parseDoctype() throws IOException {

int nesting = 1;

while (true) {
            int i = read();
            switch (i) {

case -1 :
                    exception(UNEXPECTED_EOF);

case '<' :
                    nesting++;
                    break;

case '>' :
                    if ((--nesting) == 0)
                        return;
                    break;
            }
        }
    }

/* precondition: &lt;/ consumed */

private final void parseEndTag() throws IOException {

read(); // '<'
        read(); // '/'
        name = readName();
        if (depth == 0 && !relaxed)
            exception("element stack empty");
            
        if (name.equals(elementStack[depth-1]))
                depth--;
        else if (!relaxed)
            exception("expected: " + elementStack[depth]);
        skip();
        read('>');
    }

private final int peekType() {
        switch (peek0) {
            case -1 :
                return END_DOCUMENT;
            case '&' :
                return ENTITY_REF;
            case '<' :
                switch (peek1) {
                    case '/' :
                        return END_TAG;
                    case '[' :
                        return CDSECT;
                    case '?' :
                    case '!' :
                        return LEGACY;
                    default :
                        return START_TAG;
                }
            default :
                return TEXT;
        }
    }

private static final String[] ensureCapacity(
        String[] arr,
        int required) {
        if (arr.length >= required)
            return arr;
        String[] bigger = new String[required + 16];
        System.arraycopy(arr, 0, bigger, 0, arr.length);
        return bigger;
    }

/** Sets name and attributes */

private final void parseStartTag() throws IOException {

read(); // <
        name = readName();
        elementStack = ensureCapacity(elementStack, depth + 1);
        elementStack[depth++] = name;

while (true) {
            skip();

int c = peek0;

if (c == '/') {
                degenerated = true;
                read();
                skip();
                read('>');
                break;
            }

if (c == '>') {
                read();
                break;
            }

if (c == -1)
                exception(UNEXPECTED_EOF);

String attrName = readName();

if (attrName.length() == 0)
                exception("attr name expected");

skip();
            read('=');
            
                                
            skip();
            int delimiter = read();

if (delimiter != '/'' && delimiter != '"') {
                if (!relaxed)
                    exception(
                        "<"
                            + name
                            + ">: invalid delimiter: "
                            + (char) delimiter);

delimiter = ' ';
            }

int i = (attributeCount++) << 1;

attributes = ensureCapacity(attributes, i + 4);

attributes[i++] = attrName;

int p = txtPos;
            pushText(delimiter);

attributes[i] = pop(p);

if (delimiter != ' ')
                read(); // skip endquote
        }
    }

/** result: isWhitespace; if the setName parameter is set,
    the name of the entity is stored in "name" */

public final boolean pushEntity() throws IOException {

read(); // &

int pos = txtPos;

while (!eof && peek0 != ';')
            push(read());

String code = pop(pos);

read();

if (code.length() > 0 && code.charAt(0) == '#') {
            int c =
                (code.charAt(1) == 'x'
                    ? Integer.parseInt(code.substring(2), 16)
                    : Integer.parseInt(code.substring(1)));
            push(c);
            return c <= ' ';
        }

String result = (String) entityMap.get(code);
        boolean whitespace = true;

if (result == null)
            result = "&" + code + ";";

for (int i = 0; i < result.length(); i++) {
            char c = result.charAt(i);
            if (c > ' ')
                whitespace = false;
            push(c);
        }

return whitespace;
    }

/** types:
    '<': parse to any token (for nextToken ())
    '"': parse to quote
    ' ': parse to whitespace or '>'
    */

private final boolean pushText(int delimiter)
        throws IOException {

boolean whitespace = true;
        int next = peek0;

while (!eof
            && next != delimiter) { // covers eof, '<', '"'

if (delimiter == ' ')
                if (next <= ' ' || next == '>')
                    break;

if (next == '&') {
                if (!pushEntity())
                    whitespace = false;

}
            else {
                if (next > ' ')
                    whitespace = false;

push(read());
            }

next = peek0;
        }

return whitespace;
    }

//--------------- public part starts here... ---------------

public XmlReader(Reader reader) throws IOException {
        this.reader = reader;

peek0 = reader.read();
        peek1 = reader.read();

eof = peek0 == -1;

entityMap = new Hashtable();
        entityMap.put("amp", "&");
        entityMap.put("apos", "'");
        entityMap.put("gt", ">");
        entityMap.put("lt", "<");
        entityMap.put("quot", "/"");

line = 1;
        column = 1;
    }

public void defineCharacterEntity(
        String entity,
        String value) {
        entityMap.put(entity, value);
    }

public int getDepth() {
        return depth;
    }

public String getPositionDescription() {

StringBuffer buf =
            new StringBuffer(
                type < TYPES.length ? TYPES[type] : "Other");

buf.append(" @" + line + ":" + column + ": ");

if (type == START_TAG || type == END_TAG) {
            buf.append('<');
            if (type == END_TAG)
                buf.append('/');

buf.append(name);
            buf.append('>');
        }
        else if (isWhitespace)
            buf.append("[whitespace]");
        else
            buf.append(getText());

return buf.toString();
    }

public int getLineNumber() {
        return line;
    }

public int getColumnNumber() {
        return column;
    }

public boolean isWhitespace() {
        return isWhitespace;
    }

public String getText() {

if (text == null)
            text = pop(0);

return text;
    }

public String getName() {
        return name;
    }

public boolean isEmptyElementTag() {
        return degenerated;
    }

public int getAttributeCount() {
        return attributeCount;
    }

public String getAttributeName(int index) {
        if (index >= attributeCount)
            throw new IndexOutOfBoundsException();
        return attributes[index << 1];
    }

public String getAttributeValue(int index) {
        if (index >= attributeCount)
            throw new IndexOutOfBoundsException();
        return attributes[(index << 1) + 1];
    }

public String getAttributeValue(String name) {

for (int i = (attributeCount << 1) - 2;
            i >= 0;
            i -= 2) {
            if (attributes[i].equals(name))
                return attributes[i + 1];
        }

return null;
    }

public int getType() {
        return type;
    }

public int next() throws IOException {

if (degenerated) {
            type = END_TAG;
            degenerated = false;
            depth--;
            return type;
        }

txtPos = 0;
        isWhitespace = true;

do {
            attributeCount = 0;

name = null;
            text = null;
            type = peekType();

switch (type) {

case ENTITY_REF :
                    isWhitespace &= pushEntity();
                    type = TEXT;
                    break;

case START_TAG :
                    parseStartTag();
                    break;

case END_TAG :
                    parseEndTag();
                    break;

case END_DOCUMENT :
                    break;

case TEXT :
                    isWhitespace &= pushText('<');
                    break;

case CDSECT :
                    parseLegacy(true);
                    isWhitespace = false;
                    type = TEXT;
                    break;

default :
                    parseLegacy(false);
            }
        }
        while (type > TEXT
            || type == TEXT
            && peekType() >= TEXT);

isWhitespace &= type == TEXT;

return type;
    }

//-----------------------------------------------------------------------------
    // utility methods to mak XML parsing easier ...

/**
     * test if the current event is of the given type and if the
     * name do match. null will match any namespace
     * and any name. If the current event is TEXT with isWhitespace()=
     * true, and the required type is not TEXT, next () is called prior
     * to the test. If the test is not passed, an exception is
     * thrown. The exception text indicates the parser position,
     * the expected event and the current event (not meeting the
     * requirement.
     *
     * <p>essentially it does this
     * <pre>
     *  if (getType() == TEXT && type != TEXT && isWhitespace ())
     *    next ();
     *
     *  if (type != getType
     *  || (name != null && !name.equals (getName ())
     *     throw new XmlPullParserException ( "....");
     * </pre>
     */
    public void require(int type, String name)
        throws IOException {

if (this.type == TEXT && type != TEXT && isWhitespace())
            next();

if (type != this.type
            || (name != null && !name.equals(getName())))
            exception("expected: " + TYPES[type] + "/" + name);
    }

/**
     * If the current event is text, the value of getText is
     * returned and next() is called. Otherwise, an empty
     * String ("") is returned. Useful for reading element
     * content without needing to performing an additional
     * check if the element is empty.
     *
     * <p>essentially it does this
     * <pre>
     *   if (getType != TEXT) return ""
     *    String result = getText ();
     *    next ();
     *    return result;
     *  </pre>
     */

public String readText() throws IOException {

if (type != TEXT)
            return "";

String result = getText();
        next();
        return result;
    }
}

使用方法很简单，我在这里抛砖引玉的给出一个sample，大家根据应用自己发挥吧！但是我还没有发现其能够解析内嵌element。

xml文件内容如下：

<?xml version="1.0" encoding="GBK"?>
<dataList name="list01">
        <data name="data01" type="String" value="bestMan"/>
        <data name="data02" type="String" value="bestMan2"/>
</dataList>

代码片断如下：

/读取xml文件，并显示其内容
        public void readXml() throws Exception{
                InputStream ins=this.getClass().getResourceAsStream("/test.xml");
                if (ins==null){
                        System.out.println("null InputStream");
                }
                else{
                        InputStreamReader reader=new InputStreamReader(ins);
                        xmlReader=new XmlReader(reader);
                        xmlReader.next();
                        xmlReader.require(XmlReader.START_TAG,"dataList");
                        System.out.println("dataList name attribute="+xmlReader.getAttributeValue("name"));
                        
                        while(xmlReader.next()!=XmlReader.END_DOCUMENT){
                                try{
                                        System.out.println(xmlReader.getPositionDescription());
                                        xmlReader.require(XmlReader.START_TAG,"data");
                                        System.out.println(xmlReader.getPositionDescription());
                                        xmlReader.next();
                                        System.out.println("data name="+xmlReader.getAttributeValue("name")
                                                +" type="+xmlReader.getAttributeValue("type")
                                                +" value="+xmlReader.getAttributeValue("value"));
                                        
                                        System.out.println(xmlReader.getPositionDescription());
                                        xmlReader.require(XmlReader.END_TAG,"data");
                                }
                                catch(Exception e){
                                        xmlReader.require(XmlReader.END_TAG,"dataList");
                                }
                        }
                        
                                                
                }
        }